On 15.01.19 16:38, Christoph Hellwig wrote:
> On Mon, Jan 14, 2019 at 01:58:54PM +0100, David Hildenbrand wrote:
>> Nothing major changed since the last version. I would be happy about
>> additional ACKs. If there are no further comments, can this go via the
>> mm-tree in one chunk?
>>
>> I was recently going over all users of PG_reserved. Short story: it is
>> difficult and sometimes not really clear if setting/checking for
>> PG_reserved is only a relict from the past. Easy to break things. I
>> guess I now have a pretty good idea wh things are like that
>> nowadays and how they evolved.
>
> Any reason you skipped
>
> drivers/gpu/drm/drm_pci.c:drm_pci_alloc()
>
> and
>
> drivers/gpu/drm/drm_scatter.c:drm_legacy_sg_alloc()
>
> which both look completely bogus as-is?
>
> In fact we should probably just try to kill them off as they have
> very few users left.
>
I actually have patches lying around for these, however excluded them
for now as I was not 100% sure about the implications ("Easy to break
things").
Are you sure nobody in the system (especially somebody who does a
PageReserved()) relies on this to identify e.g. MMIO pages? (e.g.
ioremap on some archs, KVM code)
(I am by far no DRM expert, but when I hear DMA, I tend to be careful)
--
Thanks,
David / dhildenb
